Hip fracture represents a problem of nearly crisis proportions in the U.S., with over 250,000 annual fractures and costs in the range of $7 to 10 billion. Despite the fact that hip fracture represents a structural failure, the biomechanics of osteoporotic fracture risk have received relatively little attention. The fundamental goal of this ongoing research program is to understand the biomechanics of hip fracture risk and, in particular, to resolve current uncertainties over the relative importance of age-related bone loss and severity of trauma in the etiology of hip fracture. We also hope to develop accurate densitometric predictors of hip fracture risk which can be implemented in a clinical setting. In the previous funding period we have shown that the energy required to fracture the cadaveric hip in-vitro is only one-sixteenth the potential energy available in a simple fall from standing height, a finding which may help explain the failure of previous densitometric measures to separate hip fracture patients from non- fractured controls. Yet the potentially dominant role of fall mechanics leaves unexplained why less than two percent of falls among the elderly actually result in hip fracture, why hip fracture incidence among young adults is so much lower than among the elderly, and why hip fracture is relatively uncommon among obese individuals. We suspect that the epidemiologic evidence can be explained in part by age-related differences in femoral strength between young adults and the elderly, to influences of loading rate and direction, and to the force attenuation and energy absorption characteristics of soft tissues overlying the hip. Our first aim therefore is to determine, through cadaveric loading experiments and finite element models, the relative importance to hip fracture risk of density, loading direction and rate, and soft tissue energy absorption. Our second aim relates to the development of new densitometric predictors of hip fracture risk based on Quantitative Digital Radiography (QDR), a densitometric modality that has emerged as the method of choice for osteoporotic fracture risk assessment. During the previous funding period, we reported a new QCT-based predictor of hip fracture risk which provides highly significant correlations (R2 = 0.93, p < 0.00001) with in-vitro fracture loads in the age range 50 to 90 years. This trochanteric index provides much improved correlations when compared against the Ward's triangle site which is commonly used clinically. Our proposed experiments will extend these results to QDR and to a broader range of from 15 to 90. Our third aim is to test the discriminatory capability of these optimized QDR-based fracture risk predictors in a prospective densitometric study of elderly fallers with and without a hip fracture. By accounting for fall severity and trochanteric soft tissues in these populations, we expect dramatic improvements in the densitometric separation between fallers who fracture and fallers who do not. We also expect these experiments to help resolve current uncertainties over the relative importance of bone loss and fall severity in the etiology of hip fracture. These findings should then allow a more informed assessment of the potential reduction in hip fracture rates from intervention efforts aimed to modulating bone loss, at decreasing the incidence of falls, or at reducing the injury potential of the falls that do occur.